Add storage device configuration for SAC policy and replay buffer

- Introduce `storage_device` parameter in SAC configuration and training settings
- Update learner server to use configurable storage device for replay buffer
- Reduce online buffer capacity in ManiSkill configuration
- Modify replay buffer initialization to support custom storage device
This commit is contained in:
AdilZouitine
2025-03-04 13:22:35 +00:00
committed by Michel Aractingi
parent ae51c19b3c
commit bb69cb3c8c
3 changed files with 11 additions and 5 deletions

View File

@@ -20,6 +20,9 @@ training:
grad_clip_norm: 10.0
lr: 3e-4
storage_device: "cpu"
eval_freq: 2500
log_freq: 10
save_freq: 2000000
@@ -30,7 +33,7 @@ training:
online_steps_between_rollouts: 1000
online_sampling_ratio: 1.0
online_env_seed: 10000
online_buffer_capacity: 1000000
online_buffer_capacity: 200000
online_buffer_seed_size: 0
online_step_before_learning: 500
do_online_rollout_async: false